Pallet repacking cut by 70%.
Efficiency up over 30%.
A 3D camera and projector replace a weight-based scale, so packers get real-time placement guidance instead of manually rearranging pallets after the fact.
repacking
efficiency gain
utilisation
productivity
Client
A major retail chain specialising in drugstore products, cosmetics, health and wellness, household items and beauty products. It runs three high-throughput distribution centres, all already operating near capacity.
Goal
Improve pallet packing across the three centres: increase picking speed while maintaining accuracy, reduce packaging errors, and fix pallet spacing inefficiencies, all without disrupting operations that were already running flat out.
Every fix for one bottleneck made another one worse.
Package detection relied on an industrial scale fitted to each robot. Confirming a placement meant taring the scale before and after every item, a check that cost 3-5 seconds per package even while the system ran at maximum OEE. It was already the ceiling on throughput.
The single-track monorail compounded the problem: robots queued behind one another with no way to overtake, so congestion spiked during peak promotions. Meanwhile packers had no visibility of picks happening in other aisles, so they rearranged loads like Tetris to make loads fit. It still only got pallets to around 75% capacity.
Three constraints were non-negotiable:
- 3-5 seconds per item just to tare the scale
- Single-track monorail: no overtaking, so one slow robot queues the rest
- Manual pallet Tetris still capped utilisation at around 75%
A camera and a projector replaced the scale.
The weight-based system was swapped for Vision AI: a 3D camera reads each package’s dimensions as it lands on the pallet, and a deep learning model confirms the detection. That cut detection time from 3-5 seconds to around 1 second per item, with no compromise on safety standards. Faster detection also let multiple packers work the same pallet at once, easing the single-file dependency that caused monorail queuing.
Packers stopped playing Tetris with pallets. The projector already knew where everything went.
A projector mounted over the packing station handles the spacing problem. An AI model reads the preloaded picklist and calculates the ideal location for each package, then projects that layout directly onto the pallet in real time. Packers place items where the projector shows, with no need to rearrange loads afterwards.
Both changes retrofit onto the existing monorail and packer workflow. There was no infrastructure overhaul, no change to safety checks, just faster detection and a projected layout doing the planning that packers used to do by hand.
Less repacking, more pallets per shift.
Picking and packing efficiency improved by over 30% and pallet repacking fell by 70%, letting the retailer complete more pallets per shift across all three distribution centres. Fewer pallets per vehicle also cut CO2 emissions, adding a sustainability gain to the efficiency one. None of it required a hardware overhaul: a camera and a projector did what a scale and a Tetris-playing packer used to do, only faster.
* Case studies reflect work undertaken by our Heads of AI either during their tenure with Head of AI or in prior roles before they were part of the Head of AI network; they are provided for illustrative purposes only and are based on conversations with our Heads of AI.
More case studies.
Your biggest pain point.
Fixed in 14 days. 50% off.
This started with one conversation. Book a 30 minute brainstorm call — we’ll plan your first AI project together and issue your 50% discount code. No payment today.
*Case studies reflect work undertaken by our Heads of AI either during their tenure with Head of AI or in prior roles before they were part of the Head of AI network; they are provided for illustrative purposes only and are based on conversations with our Heads of AI.